an autonomous robot equipped with the gps … · referred to as astm d6433 [12] for roads and...

141

AN AUTONOMOUS ROBOT EQUIPPED WITH THE GPS VIRTUAL

REFERENCE STATION (VRS) SYSTEM TO PERFORM

PAVEMENT DISTRESS SURVEYS

Jia-Ruey Chang Associate Professor

Department of Civil Engineering

MingHsin University of Science & Technology

No. 1, Hsin-Hsing Road, Hsin-Chu 304, Taiwan

(T) 886-3-5593142-3295; (F) 886-3-5573718

[email protected]

Peter M. Liu Postdoctoral Researcher


National Taiwan University

No. 1, Sec. 4, Roosevelt Road, Taipei 106, Taiwan

(T) 886-2-33664346; (F) 886-2-2368-8213

[email protected]

Shih-Chung Kang Assistant Professor




(T) 886-2-33664346; (F) 886-2-2368-8213

[email protected]

Shang-Hsien Hsieh Professor




(T) 886-3-2-33664313; (F) 886-2-2368-8213

[email protected]

Tsun-Cheng Huang Master Student


MingHsin University of Science & Technology

No. 1, Hsin-Hsing Road, Hsin-Chu 304, Taiwan

(T) 886-3-5593142-3281; (F) 886-3-5573718

[email protected]

Ping-Hung Lin Master Student




(T) 886-2-33664346; (F) 886-2-2368-8213

[email protected]

ABSTRACT

Various pavement distresses (cracking, pothole, manhole, patching, etc.) that exist randomly across pavements deteriorate

the quality of the pavement structure and need to be immediately dealt using proper maintenance and rehabilitation (M&R)

strategies. Traditionally pavement distress surveys are performed using manually operated or driven equipment, and hence

are very labour-intensive, time-consuming and lack intelligent detection methods. In this study, an autonomous robot

platform was pioneered to conduct pavement distress surveys. For accurate data collection and robot mobility on broad

pavement, a positioning system with high accuracy and a short initialization time is needed. The main objective of this

142

study is to integrate Leica’s GPS VRS system with a P3-AT robot to conduct pavement distress surveys effectively.

Pavement distresses can be located to the nearest centimeter. Through field tests, the feasibility of this architecture was

examined.

KEYWORDS

Pavement distress survey, Autonomous robot, Global positioning system (GPS), Virtual reference station (VRS) system

1. INTRODUCTION

Pavement-management decisions are factor in the

methods of evaluating pavement conditions, predict

future performance, rank the severity of problems,

prioritize needs, and determine the required repair

levels. Current and accurate distress data (including

severity, coverage, and positioning of each distress)

are needed to document the present condition of the

pavement, to determine maintenance and rehabili-

tation (M&R) needs, and to form the basis of a

pavement management system. Traditionally, pave-

ment distress surveys are performed using a repetiti-

ve and time-consuming procedure; it uses human-

operated or driven equipment, and hence are very

labour-intensive, time-consuming and lack intelli-

gent detection methods. It is very difficult to collect

pavement distress data safely, accurately, and intelli-

gently on broad pavements. To solve these problems,

this study pioneered the application of an autono-

mous robot equipped with a high-accuracy posi-

tioning system to pavement distress surveying by

exploiting robots’ features of mobility, precision, and

autonomy from human control to save on time and

cost. The robot is able to search for pavement

distresses, capture distress images, and plan the upco-

ming motion in real-time. This allows engineers to

collect pavement distress data in a more effective way.

Global Positioning System (GPS) Real-Time Kine-

matic (RTK) positioning is becoming increasingly

important for many precise GPS applications, with

examples including surveying, construction, pre-

cision farming and high accuracy Geographic

Information Systems (GIS). Traditionally, a user

receiver requires a reference station within 10 km to

ensure centimeter-level accuracy. Recently, multiple

reference station networks have been installed in

many countries to overcome the limitations of

standard RTK systems. The concept of the Virtual

Reference Station (VRS) is one of the more feasible

approaches for relaying network correction

information to the network RTK users. For

accurately positioning the robot on pavements, a

positioning system with high accuracy (centimeter-

level), Leica’s GPS VRS system, was equipped on

the P3-AT autonomous robot. In this study, a

demonstration was performed on an in-situ

pavement to assess the feasibility of this architecture

for distress data collection.

2. GPS VIRTUAL REFERENCE STATION

(VRS) SYSTEM

RTK positioning in conjunction with GPS can reach

a high accuracy (to the centimeter-level) and is one

of the most widely used surveying techniques today.

But, RTK has some drawbacks such as the effects of

the troposphere and ionosphere which result in

systematic errors in raw data; the distance between a

rover (mobile) receiver and its reference (base)

station has to be quite short (less than 10 km) in

order to work efficiently. Also radio communication

problems arise in urban areas.

Recently, the use of the VRS, so-called network

RTK, concept has been proposed by many

researchers as a more feasible approach for relaying

network correction information to the network RTK

users [1-5] and has proven an efficient technology

for high accuracy GPS positioning over the last few

years. This approach does not require an actual

physical reference station. Instead, it allows the user

to access data of a non-existent virtual reference

station at any location within the network coverage

area and allows modelling the systematic errors to

provide the possibility of an error reduction. VRS

network operation is shown in Figure 1 [6]. The

advantages of the VRS technique follow.

• increase the allowable distance between the

rover receiver and the reference station;

• remove large portions of ionospheric and

geometric errors through network corrections;

143

Figure 1. VRS Network Operation [6]

• increase the reliability and productivity of

ambiguity resolution and the positioning

accuracy of rovers;

• reduce the initialization time of RTK.

The VRS approach is flexible: users can utilize their

current receivers and software without any specific

software to manage corrections from a series of

reference stations. Users within the reference station

network can operate reliably at greater distances

without a degradation in accuracy. However, it is

necessary to provide a reliable data communication

link for the transmission of VRS data from a control

centre to a user receiver [7]. There are several

practical approaches to transfer the VRS data to

users in real-time, such as TV audio data broadcast,

GSM, Internet, etc [3, 8]. In this study, a General

Packet Radio Service (GPRS) is utilized as the data

transmission device to position the robot on

pavements. GPRS provides an “always-on” service,

and is thus preferable to GSM. It provides a stable

and reliable connection with latencies of less than

one second. Figure 2 shows the architecture for

linking the control centre with the user receiver

using GPRS [7]. The PDA in Figure 2 is replaced

with an onboard laptop on the autonomous robot in

this study.

Figure 2. The Architecture of Internet-based VRS System

via GPRS [7]

3. AN AUTONOMOUS ROBOT:

MOBILEROBOTS P3-AT

Being powerful, easy-to-use, reliable, and flexible,

the Pioneer 3-AT (P3-AT) used in this study is a

highly versatile all-terrain robotic platform

particularly suited for pavement inspections. Figure

3 shows the P3-AT.

3.1.1.1. Introduction to P3-AT

The P3-AT is manufactured by MobileRobots Inc.

and is equipped with a control laptop, onboard Pan-

Tilt-Zoom (PTZ) camera system, Ethernet-based

communications, SICK laser, eight forward and

eight rear sonars which sense obstacles from 15 cm

to 7 m, and other autonomous functions. P3-AT’s

powerful motors and four robust wheels can reach

speeds of 0.8 m/sec and carry a payload of up to 30

kg. P3-AT can climb steep 45% grades and sills of 9

cm. P3-AT uses 100-tick encoders with inertial

correction recommended for dead reckoning to

compensate for skid steering. Its sensing extends far

beyond the ordinary with laser-based navigation

options, integrated inertial correction to compensate

for slippage, bumpers, gripper, vision, stereo

rangefinders, compass and a rapidly growing suite of

other options [9].

The bare P3-AT base with included Advanced

Robotics Interface Application (ARIA) software has

the ability to [9]:

• wander randomly on pavements;

• drive controlled by keys or joystick;

• plan survey paths with gradient navigation on

pavements;

• display a pavement spatial map of its sonar

and/or laser readings;

• localize using sonar or laser upgrade;

• communicate sensor and control information

relating sonar, motor encoder, motor controls,

user I/O, and battery charge data;

• test pavement survey activities quickly with

ARIA API from C++ programs;

• simulate pavement survey behaviours offline

with the simulator that accompanies each deve-

lopment environment.

144

Figure 3. MobileRobots P3-AT

•

3.2.1.2. P3-AT Vision System

The Pan-Tilt-Zoom (PTZ) system on the P3-AT

provides an integrated robotic camera whose

movement can be easily controlled via ARIA

development software. This camera can be

configured for pavement distress survey purposes

with specific vision routines [10].

The PTZ robotic camera system is fully integrated

with ARCOS (robot hardware and operating system

software) and ARIA software plug-ins for easy

operation. The system includes:

• 26 times (at infinite) optical, 12X digital zoom

for wide range of applications;

• high-speed auto-focus lens for responsiveness;

• +/- 100 degree pan at up to 90 degrees/sec;

• -30 to +90 degree tilt at up to 70 degrees/sec;

• connections to plug into MobileRobots

microcontroller RS232C serial port; and,

• all NTSC functions.

ARCOS and ARIA command sets include:

• pan/tilt to x/y degrees (absolute position);

• pan/tilt x/y degrees (relative position);

• 1X - 26X zoom.

4. GPS VRS SYSTEM INTRGRATED WITH

P3-AT TO PAVEMENT DISTRESS SURVEYS

A pavement distress survey was conducted to

monitor the present pavement condition and

determine the appropriate M&R activities. The

common distress types include alligator cracking,

longitudinal cracking, transverse cracking, block

cracking, rutting, and patching, etc. Toward the end

of the 1970s, Darter and Shahin [11] developed the

Pavement Condition Index (PCI), which has been

referred to as ASTM D6433 [12] for roads and

parking Lots and ASTM D5340 [13] for airfield

pavement. Manual and automatic approaches are the

two major ways, adopted all over the world, to

perform a pavement distress survey. The manual

approach is conducted through visual observation.

The severity, coverage, and positioning of each

distress on one pavement unit are identified and

recorded using pen and paper or a personal digital

assistant (PDA) equipped with GPS. The PCIs can

be computed immediately using the PDA or

manually during post-processing. The automatic

approach is to first capture pavement images with

GPS information, and then the severity and coverage

of each distress are digitally identified using image

processing algorithms. However, present image

processing merely focuses on linear cracking and

potholes thus could not replace the manual approach.

Significant developments have taken place in

automating distress measurement. There are many

inertial profilers that automate the complete distress

data collection. More literature can be found at [14,

15].

Most distress surveys on broad pavements are

performed using a repetitive and time-consuming

procedure that uses human-operated instruments,

hence are very labour-intensive, time-consuming

and lack the application of intelligent technologies.

To solve these problems, this study proposes the

idea of applying P3-AT equipped with an Internet-

based VRS system via GPRS to pavement distress

surveys by utilizing the robot’s features of mobility,

high accuracy, and time and labour efficiencies. P3-

AT will search pavement distresses, capture distress

images with positioning information in real time,

and plan the upcoming motion in real-time. This

PTZ Camera

System SICK Laser

Rangefinder

Sonar Sensors

145

helps us collect pavement distress data in a more

effective way.

4.1.1.3. GPS VRS System Operation

Software, SpiderNET, at the control-centre server

located at the Control Signal Company Limited has

been written to generate and broadcast VRS data via

the Internet. The software opens the ports and waits

for the P3-AT connection. There is a login procedure

required to gain access to the data. The software

handles incoming logins and generates the VRS data

for P3-AT according to P3-AT’s approximate

position.

On the robot side, P3-AT is equipped with a Leica

SR530 receiver (Figure 4) capable of performing

RTK positioning and is connected to a GPRS-

enabled laptop via the RS232 port (COM 1). Client

software, Ntrip Client, was installed on the PC

environment, as shown in Figure 5. After filling the

IP port, User-ID, and password in the Ntrip Client,

the laptop establishes communication with the

receiver through another RS232 port (COM 2), as

shown in Figure 6. P3-AT can connect to the server

of the control centre through GPRS at its location.

The positioning quality can reach to within 0.01 m

(centimeter-level), as shown in Figure 7.

Figure 4. The Profile of P3-AT for Pavement Distress

Surveys

Figure 5. Ntrip Client Software

Figure 6. VRS Data Access by Ntrip Client

Figure 7. P3-AT Positioning Quality Reaches to 0.01 m

(centimeter-level)

Leica SR530

GPS Antenna

146

4.2.1.4. Pavement Distress Surveys

The distress images were automatically captured by

the onboard PTZ camera system and recorded with

positioning information of centimeter-level quality

in real-time, as shown in Figure 8. The image shown

in Figure 8 was captured at 07:44:55am in 2008-3-

23. The cracking on the pavement occurs at North

Latitude 2501.1342162 (WGS 1984) and East

Longitude 12132.3516293 (WGS 1984), according

to the data superimposed on the image. However, at

present, the distress recognition and PCIs

computation were conducted manually during the

post-processing. Development of an overall

automatic process is ongoing.

Figure 8. A Cracking Image Captured by P3-AT with

Positioning Information (centimeter-level)

5. CONCLUSIONS

For correct pavement data collection and mobility of

a robot on broad pavements, a positioning system

that possesses high accuracy, allows for a longer

distance between the GPS receiver and the reference

station and is also free of radio communication

problems in urban areas is required. The Internet-

based VRS system is a real-time, centimeter-level

service for construction, rapid surveying, and GIS.

In the study, Leica’s GPS VRS system is integrated

with an autonomous robot P3-AT and applied

successfully in the pavement distress surveys for

high accuracy positioning acquisition via GPRS. It

can also be used in related pavement inspection

applications including inspections for longitudinal

evenness (roughness) and transversal evenness

(rutting). This helps us to collect pavement condition

data in a more effective way.

6. ACKNOWLEDGEMENTS

7.The authors would like to thank the National

Science Council of Taiwan for their financial

support provided under the project of NSC 96-2628-

E-159-002-MY3, and the technical support received

from the Control Signal Company Limited in

Taiwan.

REFERENCES

[1] Wanninger, L. (1997) Real-time differential GPS-

error modelling in regional reference station

networks, Proc. of the IAG Scientific Assembly, Rio

de Janeiro, September, IAG-Symposia 118, Springer

Verlag, 86–92.

[2] Marel, H. (1998) Virtual GPS reference stations in

the Netherlands, Proc. 11th Int. Tech. Meeting of the

Satellite Division of the U.S. Inst. of Navigation,

ION GPS-98, Nashville, TN, September 15–18, 49–

58.

[3] Vollath, U., Buecherl, A., Landau, H., Pagels, C., &

Wager, B. (2000) Multi-base RTK positioning using

virtual reference stations, Proc. 13th Int. Tech.

Meeting of the Satellite Division of the U.S. Inst. of

Navigation, ION GPS-2000, Salt Lake City,

September 19–22, 123–131.

[4] Cannon, M.E., Lachapelle, G., Fortes, L.P., Alves,

P., & Townsend, B. (2001) The use of multiple

reference station VRS for precise kinematic

positioning, Proc. of the Japan Institute of

Navigation, GPS Symposium 2001, Tokyo,

November 14–16, 29–37.

[5] Euler, H.J., Keenan, C.R., Zebhauser, B.E., &

Wübbena, G. (2001) Study of a simplified approach

in utilizing information from permanent reference

station arrays, Proc. 14th International Technical

Meeting of the Satellite Division of the Institute of

Navigation, Salt Lake City, USA, ION GPS-2001,

September 11–14, 379–391.

[6] Trimble Virtual Reference Station (VRSTM) System

(2005), Versatile RTK positioning infrastructure for

local and regional applications, Retrieved March 10,

2008, from http://trl.trimble.com/docushare/dsweb/

Get/Document-11465/VRS_Spec_Sheet.pdf

[7] Hu, G.R., Khoo, V.H.S., Goh, P.C., & Law, C.L.

(2002) Internet-based GPS VRS RTK Positioning

with a Multiple Reference Station Network, Journal

147

of Global Positioning Systems, Vol. 1, No. 2, 113–

120.

[8] Liu, Z. & Gao, Y. (2001) Research toward wireless

Internet-based DGPS, Proceedings of KIS 2001,

Banff, Canada, June 5–8, 461–469.

[9] MobileRobots Inc. (2008) The high performance all-

terrain robot - P3-AT, Retrieved January 21, 2008,

from http://www. activrobots.com/ROBOTS/

p2at.html

[10] MobileRobots Inc. (2008) Robotic Pan-Tilt-Zoom

camera systems, Retrieved Feb. 13, 2008, from

http://www.activrobots.com/ACCESSORIES/

ptzvideo.html#robocam

[11] Darter, M.I. & Shahin, M.Y. (1980) Pavement

rehabilitation: Identifying the need, Journal of

Transportation Engineering, Vol. 106, No. 1, 1–10.

[12] ASTM D6433 (2003) Standard Practice of Roads

and Parking Lots Pavement Condition Index

Surveys.

[13] ASTM D5340 (2004) Standard Test Method for

Airport Pavement Condition Index Surveys.

[14] Wang, K.C.P. (2003) Transportation Research

Circular: Automated Imaging Technologies for

Pavement Distress Survey, Committee A2B06,

Transportation Research Board, National Research

Council, Washington, D.C.

[15] Wang, K.C.P. & Li, X. (1999) Use of digital camera

for pavement surface distress survey, Transportation

Research Record 1675, Transportation Research

Board, National Research Council, Washington,

D.C., 91–97.

an autonomous robot equipped with the gps … · referred to as astm d6433 [12] for roads and...

Documents